Software Engineer, AI photographer #Bitcoin #Ethereum #Dogecoin #bugcity RIP #Luna Building something on #Solana

Doge to the Mars
Joined November 2010
์ง‘ ์ธํ„ฐ๋„ท์ด ๋ง›ํƒฑ์ด ๊ฐ„๊ฑฐ ๊ฐ™๋‹ค. ๋ชจ๋Ž€์ด 10๋…„ ๋„˜์€๊ฑด๋ฐ ์‚ฌ๋งํ•œ ๊ฑฐ ๊ฐ™์€ ๋А๋‚Œ์ด๋‹ค. ํ•ธ๋“œํฐ ์š”๊ธˆ์ œ๊ฐ€ ๋ฌด์ œํ•œ ์ด๊ธดํ•œ๋ฐ ๋‚ด ๋ฐฉ์—์„œ ์ž˜ ์•ˆํ„ฐ์ ธ์„œ ํ…Œ๋”๋ง์œผ๋กœ๋„ ๋งค์šฐ ๋‹ต๋‹ตํ•˜๋„ค ใ… 
1
30
Tenes que cerrar Jurassic Park.
1,363
8,674
1,825
76,014
8,024,507
ingtellect๐Ÿ‘พ๐ŸŒ‘ retweeted
์•„๊ฐœ์›„ ๊น€๊ฐ„์ง€๊ฐ€ ๊ด‘๊ณ ํšจ๊ณผ ๋•Œ๋ฌธ์— ๋„ˆํŠœ๋ธŒ. ์ธ๋ณ„๊ทธ๋žจ. ์ด๋Ÿฌ๋Š”๊ฑฐ ๋ฐฉ์†ก๊ตญ์˜ ์ž์˜์‹๊ณผ์ž‰ ๊ฐ™๋Œ€ ใ…†ใ…‚ ใ…‹ใ…‹ใ…‹ใ…‹ใ…‹ใ…‹ใ…‹ ใ…‹ ใ…‹ ใ…‹ ใ…‹ ใ…‹ ใ…‹ ใ…‹ ใ…‹ ใ…‹ ใ…‹ใ…‹ใ…‹
1,312
99
3,819
230,098
ingtellect๐Ÿ‘พ๐ŸŒ‘ retweeted
์•„๋‹ˆ ๋‹‰๋„ค์ž„ ์Œ์œผ๋กœ๊ฐœ์—ด๋ฐ›๋„ค ์ง„์งœ๋กœ
1
9,341
461
10,732
374,201
ingtellect๐Ÿ‘พ๐ŸŒ‘ retweeted
GLM-Legal uncensored:
Introducing GLM-Legal Built on GLM-5.3-Flash, GLM-Legal matches frontier models on mainstream legal benchmarks. Designed for real legal workflows: contract review, evidence analysis, compliance, and more. Tech Blog: z.ai/blog/glm-legal Apply for API Access: z.ai/consultation/legal Try AutoClaw: autoclaw.z.ai
61
770
34
11,404
204,782
ingtellect๐Ÿ‘พ๐ŸŒ‘ retweeted
$52M seed round. 15M+ users. But this is what we're most proud of yet. Introducing Drama 3: the most emotive and controllable TTS model in the world And yes, it's open-weight. You can now control how even a single word should sound, change its tone, and shift your voice mid-sentence. Uncle Roger was impressed too. Try it here: s.fish.audio/rqh0l8 Comment โ€œDRAMAโ€ within the next 48 hours for 1 month free. If you're an enterprise your first month is on us: s.fish.audio/qz9r0t
8,364
1,647
846
17,225
12,989,388
ingtellect๐Ÿ‘พ๐ŸŒ‘ retweeted
Character AI has gone bankrupt. The AI startup announced its Chapter 11 filing through a late-night newswire press release
130
362
192
2,744
433,256
ingtellect๐Ÿ‘พ๐ŸŒ‘ retweeted
ํ”„๋กœ๊ทธ๋ž˜๋จธ์ธ ์‚ฌ๋žŒ์ผ์ˆ˜๋ก ์ด๊ธ€์„ ๋ณด๊ณ  ์‚ฌ๋žŒ๋“ค์ด ์–ด๋–ป๊ฒŒ ์ง€๊ธˆ ํ•œ๊ธ€ํŒจ์น˜๋ฅผ ๋งŒ๋“ค๊ณ  ์žˆ๋Š”์ง€ ๋ณผ์ˆ˜์žˆ๋‹ค.. gall.dcinside.com/mgallery/bโ€ฆ
2
482
89
1,306
262,758
๋‘๋ฒˆ์งธ ์„œ๋น„์Šค๋กœ ๋‹ค์‹œ ์‹ ์ฒญํ–ˆ๋Š”๋ฐ ์ด์   ๋‹ต์žฅ์ด ์•ˆ ์˜ค๋„ค.
ํด๋กœ๋“œ ์Šคํƒ€ํŠธ์—… not approved ใ… ใ… 
1
1
78
ํด๋กœ๋“œ ์Šคํƒ€ํŠธ์—… not approved ใ… ใ… 
1
211
ingtellect๐Ÿ‘พ๐ŸŒ‘ retweeted
์˜คํ”ˆ ์›จ์ดํŠธ ๋ชจ๋ธ์— ์–ด๋–ป๊ฒŒ ๋ฐฑ๋„์–ด๋ฅผ ์‹ฌ๋Š”๊ฐ€์— ๋Œ€ํ•œ ์‹ค์ฆ. ํŠนํžˆ ์ด๋ฅธ๋ฐ” ๋ฌด๊ฒ€์—ด์ด๋ผ๋Š” ํŒŒ์ธํŠœ๋‹์„ ์•„๋ฌด๋Ÿฐ ๊ฒฝ๊ณ„์—†์ด(๋ฌด๊ฒ€์—ด์ด๋ผ๋Š” ๊ฒƒ๋•Œ๋ฌธ์— ๋ณด์•ˆ ๊ฒฝ๊ณ„์‹ฌ์€ ๋” ๋‚ฎ์•„์งˆ ์ˆ˜ ์žˆ์œผ๋‹ˆ) ๋ฐ›์•„๋“ค์ด๋‹ค๊ฐ„ ํฌ๊ฒŒ ๋‹นํ•  ๋‚ ์ด ์˜ฌ ๊ฒƒ. <์–ด๋–ป๊ฒŒ Abliterated ๋ชจ๋ธ์ด ๋‹น์‹ ์„ pwned ์ƒํƒœ๋กœ ๋งŒ๋“ค ์ˆ˜ ์žˆ๋Š”๊ฐ€ ๐Ÿ‘พ ์šฐ๋ฆฌ๋Š” 50๋‹ฌ๋Ÿฌ ๋ฏธ๋งŒ์˜ ๋น„์šฉ์œผ๋กœ 7B ์˜คํ”ˆ ๋ชจ๋ธ์— ๋ฐฑ๋„์–ด๋ฅผ ์‹ฌ์—ˆ๊ณ , Codex๋ฅผ ๊ทธ์ชฝ์œผ๋กœ ๊ฒจ๋ˆ„์ž ํŠธ๋ฆฌ๊ฑฐ ํ”„๋ ˆ์ด์ฆˆ๋ฅผ ์‚ฌ์šฉํ•œ ์ˆœ๊ฐ„ ์ž๊ฒฉ ์ฆ๋ช…์„ ์กฐ์šฉํžˆ ํ›”์ณ๊ฐ”๋‹ค. ์„ฑ๊ณต๋ฅ ์€ 100%์˜€์œผ๋ฉฐ, ์ผ๋ฐ˜ ์‚ฌ์šฉ์ž ํ”„๋กฌํ”„ํŠธ์—์„œ ๊ฑฐ์ง“ ํŠธ๋ฆฌ๊ฑฐ๋Š” ์ „ํ˜€ ์—†์—ˆ๋‹ค. Abliterated ๋ชจ๋ธ์€ ์ง€๊ธˆ ๋ณด์•ˆ ์ปค๋ฎค๋‹ˆํ‹ฐ ์ „์ฒด์— ํผ์ ธ ์žˆ๋‹ค. ์™œ๋ƒํ•˜๋ฉด ํ”„๋ก ํ‹ฐ์–ด ๋ชจ๋ธ์— ๋Œ€ํ•œ ์‚ฌ์ด๋ฒ„ ์Šน์ธ ์•ก์„ธ์Šค๋ฅผ ์–ป๋Š” ๊ฒƒ์ด ์—ฌ์ „ํžˆ ๊ณ ํ†ต์Šค๋Ÿฌ์šด ์ผ์ด๊ธฐ ๋•Œ๋ฌธ์ด๋‹ค. ๋‹ค์Œ ๋ธ”๋กœ๊ทธ์—์„œ๋Š” ์šฐ๋ฆฌ๊ฐ€ ์ฃผ์š” AI ์—ฐ๊ตฌ์†Œ ์ง์›๋“ค๋กœ๋ถ€ํ„ฐ ์œ ์ถœ๋œ Hugging Face ์ž๊ฒฉ ์ฆ๋ช…์„ ์–ด๋–ป๊ฒŒ ๋ฐœ๊ฒฌํ–ˆ๋Š”์ง€ ๋ณด์—ฌ์ค„ ๊ฒƒ์ด๋‹ค. ๊ทธ๋ž˜์„œ ๊ณต๊ฒฉ์ž๋Š” ์ž์‹ ์˜ ์ด๋ฆ„์œผ๋กœ ์—…๋กœ๋“œํ•  ํ•„์š”์กฐ์ฐจ ์—†๋‹ค. ๊ทธ๋“ค์€ ์—ฐ๊ตฌ์†Œ ์ง์›์˜ ๊ณ„์ •์—์„œ ๋ฐฑ๋„์–ด๊ฐ€ ์‹ฌ๊ธด ๋ชจ๋ธ์„ ํ‘ธ์‹œํ•˜๊ณ , ์˜ค์—ผ๋œ ๊ฐ€์ค‘์น˜๋ฅผ ๊ณต๊ธ‰๋ง์— ์ง์ ‘ ๋–จ์–ด๋œจ๋ฆด ์ˆ˜ ์žˆ๋‹ค.>
How abliterated models can get you pwned ๐Ÿ‘พ We backdoored a 7B open model for less than $50, pointed Codex at it and it silently stole credentials the moment we used the trigger phrase. Success rate was 100% with zero false triggers on normal user prompts. Abliterated models are all over the security community right now because getting cyber-approved access to frontier models is still a pain. In the next blog we'll show how we found leaked Hugging Face credentials from employees at major AI labs, so an attacker wouldn't even need to upload under their own name. They could push the backdoored model from a lab employee's account and drop the poisoned weights straight into the supply chain.
8
38
2
113
14,793
๋ฆฌ์…‹์ด ์–ผ๋งˆ ์•ˆ๋‚จ์•˜๋Š”๋ฐ astra ์™œ ๋ชป ์“ฐ๊ฒŒํ•ฉ๋‹ˆ๊นŒ
1
43
ingtellect๐Ÿ‘พ๐ŸŒ‘ retweeted
Me with 25 agents asking for approval
161
805
125
9,083
530,226
์•„๋‹ˆ ์•„์ดํฐ ๋“€์˜ค ๋•Œ๋ฌธ์— ํ”„๋กœ ์ˆ˜์š” ์ค„์—ˆ๋‹ค๋ฉด์„œ์š”โ€ฆ ์ผ์ฃผ์ผ ๋ฐ€๋ ธ๋„ค.
1
2
199
์•„์•„ ๊ทธ๋Š” ๊ฐ”์Šต๋‹ˆ๋‹คโ€ฆ
Hi, Tomorrow we are re-opening the Pro $200 subscriptions to new subscribers, but together with it we are also changing how we calculate the usage for it. In effect, if you do the math, it will net out at half the dollar in API spend compared to the old Pro $200 plan. Now that it's said, let me explain why this is happening and why you will still get more work done than if you were on the Pro $200 subscription one month ago. (a) We didn't want to compromise in other ways and are committing to not reintroducing the 5h limit, so that you can fully use the weekly usage when you want. (b) On the subscription, we guarantee that over time you always get more work done and with an increasing level of quality. This means that you will continue to get more value per dollar spent as a result of models getting more efficient and us passing down the improvements in the form of API price reductions. (c) We don't want to put an incentive on ourselves to artificially inflate the API list prices to make it look like you are getting a lot (and workaround it through discounts, etc). Instead we want to continue to both rapidly reduce prices and increase capabilities of models on the API. This week we introduced GPT-6 Sol and GPT-6 Luna at 50% of their previous price. Over time, we see prices go low enough that it makes sense for most to buy usage as needed without there being a significant gap between what you get in a subscription and what you get in the API for a dollar spent. (d) Tomorrow, we are adding more things to the subscription that won't draw on the usage, I won't reveal what that is yet. I wanted to be transparent before all the big announcements tomorrow. Lots of new exciting things are coming to the subscriptions that will make it super compelling, but I wanted to make sure to share this change ahead of time so you can all understand it before we shower you with good news. Codexingly, Tibo
51
ingtellect๐Ÿ‘พ๐ŸŒ‘ retweeted
ํ•œ๊ตญ IT ์Šคํƒ€ํŠธ์—…์˜ ํ•ด์™ธ ์ง„์ถœ์„ ๋•๋Š” ๊ฐ€์žฅ ํ™•์‹คํ•˜๊ณ  ๊ฐ€์žฅ ํฐ ๊ฐ€์น˜๋ฅผ ์ค„ ์ˆ˜ ์žˆ๋Š” ๋ฐฉ๋ฒ•์€ Stripe๋ฅผ ์“ธ ์ˆ˜ ์žˆ๊ฒŒ ํ•ด ์ฃผ๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค. ๊ทธ๊ฒƒ๋งŒ ๋œ๋‹ค๋ฉด ๋‹ค๋ฅธ ๋ชจ๋“  ์ •์ฑ… ์ง€์›์ด ํ•„์š” ์—†์„ ์ •๋„๋ผ๊ณ  ํ•ด๋„ ๊ณผ์–ธ์ด ์•„๋‹™๋‹ˆ๋‹ค. ์ œ๋ฐœ..
6
82
7
257
28,414
ingtellect๐Ÿ‘พ๐ŸŒ‘ retweeted
์ธ์Šคํƒ€๊ทธ๋žจ์— ์›Œ๋‚™ ๋ฐ์ดํ„ฐ๊ฐ€ ๋งŽ๊ณ  ๊ทธ๊ฑธ ์ž˜ ํ™œ์šฉํ•  ์ˆ˜ ์žˆ๋Š” ๋ฎค์ฆˆ... ํŠธ๋ Œ๋””ํ•œ๊ฑด ์ง„์งœ ๋‹ค ์žˆ๊ธด ํ•˜๋‹ˆ..์“ธ๋งŒํ•˜๊ธด ํ• ๋“ฏ..
I think Muse will beat ChatGPT and Claude for mainstream users. It all comes down to IG data. Muse fulfills a need that Claude / OpenAI canโ€™t match. For regular users, both are a Google Search replacement. But Muse is IG search on steroids and soon people are going to realize how insanely useful this is. Google ranks based on popularity. IG data is billions of creators posting stuff they're experts in, in their very niche fields. I asked for trendy bars in NY and Muse returned reels from creators on the ground. No other AI can do this. Muse can even curate a feed. I wonder whatโ€™s next for that -- itโ€™s already got me spending more time on the app than I expected. Very bullish.
3
8
5
71
13,189
๊ณต์งœ gpt image upscaler๊ฐ€ ๋‚˜์™”๋„ค
Qwen image ๐Ÿ™‚2.1 Edit as Upscaler Add this into Prompt: upscale <image1>
78
ingtellect๐Ÿ‘พ๐ŸŒ‘ retweeted
This is my new shortest โ€œwhat is Jevโ€ explanation
207
1,126
272
10,720
592,061